mwaskom/seaborn

★ 13,869PythonAudience · dataComplexity · 2/5Setup · easy

Mindmap

mindmap
  root((repo))
    What it does
      Statistical charts
      Builds on matplotlib
      Works with pandas
    Chart Types
      Scatter plots
      Heatmaps
      Distribution plots
      Bar charts
    Tech Stack
      Python
      matplotlib
      pandas
      numpy
    Audience
      Data scientists
      Researchers
      Analysts
    Setup
      pip install
      conda install

mindmap root((repo)) What it does Statistical charts Builds on matplotlib Works with pandas Chart Types Scatter plots Heatmaps Distribution plots Bar charts Tech Stack Python matplotlib pandas numpy Audience Data scientists Researchers Analysts Setup pip install conda install

Click or tap to explore — scroll the page freely

Things people build with this

USE CASE 1

Plot the distribution of a dataset column with an overlaid density curve using a single seaborn function call.

USE CASE 2

Create a correlation heatmap from a pandas DataFrame to quickly spot relationships between numeric variables.

USE CASE 3

Build a multi-panel chart comparing several variables across different category groups in a dataset.

USE CASE 4

Generate scatter plots with regression trend lines for exploratory data analysis on tabular data.

Tech stack

Pythonmatplotlibpandasnumpyscipystatsmodels

Getting it running

Difficulty · easy Time to first run · 5min

In plain English

Seaborn is a Python library for making charts and graphs from data. It is built on top of a more low-level plotting library called matplotlib, and it adds a simpler, higher-level set of commands for creating statistical visualizations. The idea is that you can produce a polished chart with less code than you would need writing matplotlib instructions directly. The types of visualizations seaborn is designed for are broadly statistical in nature: things like scatter plots with trend lines, distribution plots showing how data is spread, heatmaps, bar charts with error bars, and charts that compare multiple variables at once. It works well with pandas, which is the standard Python tool for working with tabular data like spreadsheets. If you have a table of data in pandas, you can pass it directly to seaborn functions and get a chart back in a few lines. Installing seaborn is straightforward. The basic version is available through pip or conda with a single command, and it brings in numpy, pandas, and matplotlib automatically. If you want more advanced statistical features, like regression modeling or kernel density estimation, you can install an extended version that also pulls in scipy and statsmodels. Seaborn has a full documentation site with tutorials for beginners, an example gallery showing many chart types, and a detailed API reference. The project has also been published as a paper in the Journal of Open Source Software, which can be cited in academic work. This is a mature, widely used library in the Python data science world. It requires Python 3.8 or newer. Development happens on GitHub, and the project uses standard tools for testing and code style checking.

Copy-paste prompts

Prompt 1

Using seaborn, create a heatmap of the correlation matrix of a pandas DataFrame with annotated values and a blue-to-red color scale.

Prompt 2

Show me how to use seaborn's pairplot to visualize relationships between all numeric columns in my dataset, with points colored by a categorical column.

Prompt 3

I have a pandas DataFrame with monthly sales data. Use seaborn to create a bar chart with error bars comparing average sales across product categories.

Prompt 4

Write Python code using seaborn to plot a histogram with a KDE curve for a specific column, then save the figure as a high-resolution PNG.

Open on GitHub → Explain another repo

← mwaskom on gitmyhub — every repo by this author, as a profile.

Verify against the repo before relying on details.